Back to EveryPatent.com
| United States Patent |
5,251,970
|
|
Braschel
, et al.
|
October 12, 1993
|
Method of controlling the brake pressure in an anti-lock vehicle brake
system
Abstract
A method of controlling the brake pressure in an anti-lock vehicle brake
system provides for short distances to stop to be reached even with
greatly differing mass moments of inertia by calculating the pressure
reduction period of a control cycle currently under way as the product of
a first factor and a second factor. The first factor depends on the
pressure reduction period in the preceding control cycle, and the second
factor depends on the maximum re-acceleration of the wheel in the
preceding control cycle.
| Inventors:
|
Braschel; Volker (Neuwied, DE);
Seitz; Dieter (Neuwied, DE)
|
| Assignee:
|
Lucas Industries public limited company (Birmingham, GB2)
|
| Appl. No.:
|
688539 |
| Filed:
|
June 20, 1991 |
Foreign Application Priority Data
| Current U.S. Class: |
303/185 |
| Intern'l Class: |
B60T 008/32 |
| Field of Search: |
303/95,96,97,100,105,106,110
|
References Cited
U.S. Patent Documents
| 3857612 | Dec., 1974 | Bynum | 303/105.
|
| 3936941 | Feb., 1976 | Hiscox | 303/105.
|
| 4701855 | Oct., 1987 | Fennel | 303/105.
|
| 4900099 | Feb., 1990 | Braschel | 303/95.
|
| 4923255 | Mar., 1990 | Braschel et al. | 303/105.
|
| 4991910 | Feb., 1991 | Shimanuki et al. | 303/105.
|
| Foreign Patent Documents |
| 198691 | Oct., 1986 | EP.
| |
| 329071 | Aug., 1989 | EP.
| |
| 334275 | Sep., 1989 | EP.
| |
| 338414 | Oct., 1989 | EP.
| |
| 2741377 | Mar., 1978 | DE | 303/109.
|
| 01-36362 | May., 1990 | JP | 303/105.
|
| 8802709 | Apr., 1988 | WO.
| |
Other References
Western Electric Technical Digest No. 41, Jan. 1976, by S. J. Vahaviolos.
|
Primary Examiner: Graham; Matthew C.
Attorney, Agent or Firm: Oblon, Spivak, McClelland, Maier & Neustadt
Claims
What is claimed is:
1. In an anti-lock vehicle brake system wherein the rotational behavior of
a braked wheel is measured and, if a risk of locking exists, the brake
pressure is decreased and then increased in a series of successive control
cycles, a method of establishing the pressure reduction period
t.sup.n.sub.ab of a current control cycle (n) and controlling the
anti-lock vehicle brake system thereby, comprising the steps of:
storing the pressure reduction period (t.sup.n-1.sub.ab) for the
immediately preceding control cycle,
establishing a first factor corresponding to said pressure reduction
period,
storing the maximum reacceleration (peak.sub.n-1) of said wheel in the
immediately preceding control cycle,
establishing a second factor corresponding to said maximum reacceleration,
determining the pressure reduction period for said current cycle by
mathematically combining said first and second factors, and
controlling the anti-lock vehicle brake system to lower the brake pressure
on said wheel during said determined pressure reduction period.
2. The method as claimed in claim 1, characterized in that the first factor
is the pressure reduction period (t.sup.n-1.sub.ab) in the preceding
control cycle.
3. The method as claimed in claim 1, characterized in that the second
factor depends on a quotient of a predetermined value (peak-soll) and the
maximum re-acceleration (peak.sub.n-1) of the wheel measured in the
preceding control cycle (n-1).
4. The method as claimed in claim 1, characterized in that the pressure
reduction period (t.sup.n.sub.ab) of a control cycle (n) currently under
way is extended if the deceleration of the wheel become greater than
predetermined threshold values.
5. The method as claimed in claim 3, further comprising the step:
measuring a speed reduction of the wheel in the preceding control cycle
(n-1); and
wherein, the predetermined value (peak-soll) is a function of the measured
speed reduction of the wheel in the preceding control cycle (n-1).
6. The method as claimed in claim 1, characterized in that subsequent to a
pressure reduction period (t.sub.ab) the brake pressure is maintained
constant in the current control cycle (n) and thereupon the brake pressure
reduction is continued if the circumferential speed of the wheel has
diminished by a predetermined value since the end of the pressure
reduction period (t.sub.ab).
7. The method as claimed in claim 1, characterized in that the brake
pressure reduction is terminated at once if the braked wheel experiences
re-acceleration greater than a predetermined threshold value during the
control cycle (n) currently under way, prior to the end of the pressure
reduction period (t.sup.n.sub.ab).
8. The method as claimed in claim 1, characterized in that the pressure
reduction period (t.sup.n.sub.ab) of the control cycle (n) currently under
way is adjusted to equal the pressure reduction period (t.sup.n-1.sub.ab)
of the preceding control cycle (n-1) if the control deviation in the
preceding control cycle (n-1) was less than a predetermined comparative
value.
Description
FIELD OF INVENTION
The invention relates to a method of controlling the brake pressure in an
anti-lock vehicle brake system, comprising the features recited in the
preamble of claim 1.
DESCRIPTION OF THE PRIOR ART
During a braking operation under anti-lock control the rotational speed of
a braked wheel is measured and the brake pressure reduced at the
respective wheel, if the deceleration and/or slip of the wheel exceed
predetermined threshold values, so that it will again pick up speed in
order to get from the so-called instable range of the coefficient of
friction/slip curve which includes the risk of locking, into the stable
range of that curve.
Various criteria are known in the art for determining the point in time at
which to end the brake pressure reduction. Thus it was proposed, for
instance, to stop the pressure reduction when the wheel circumference
acceleration reaches a predetermined value (e.g. -1 g). According to
another known solution the pressure reduction is stopped when the wheel
circumference retardation does not become greater any more (turning
point). According to a third known solution the pressure reduction is
stopped when the decrease of the wheel circumference acceleration or
deceleration reaches a predetermined value.
These known solutions, however, raise problems particularly on roads having
a very high coefficient of friction because the so-called dead or idle
times (detecting the number of revolutions of the wheels, calculating the
valve control signals, and switching times of the magnetic valves) result
in a rather long pressure reduction or in the completion of the pressure
reduction coming too late so that distance to stop is given away
(so-called under-braking). On roads with a low coefficient of friction, on
the other hand, the solutions mentioned above provide too little pressure
reduction if it is terminated already at -1 g, for example.
It is likewise known in the art to set fixed time periods for the pressure
reduction and have them followed by a fixed holding period after which the
pressure reduction is continued. Yet this method entails a great so-called
control deviation if the first pressure reduction was insufficient to
bring the wheel back into the stable range of the coefficient of
friction/slip curve. That is true in particular with low coefficients of
friction since the duration of the first pressure reduction must be
selected such that under-braking will not occur at a high coefficient of
friction.
At high coefficients of friction (good road conditions), therefore, too
much under-braking (giving away distance to stop) can be avoided only by
completing the brake pressure reduction relatively soon. Premature
termination of the pressure reduction, however, has the disadvantage in
the prior art that the control deviations (variations of brake pressure)
are rather low at an increasing mass moment of inertia and there is a risk
that the wheels turn too long at high slip or even become locked at an
early stage.
As explained above, braking under anti-lock control depends on the mass
moment of inertia. The system composed of brake and vehicle may have
greatly varying mass moments of inertia which depend especially on whether
a light metal or steel rim is used, whether the wheel diameter is great or
small, whether the braked wheel is driven, whether the vehicle drive is
engaged or not, in which gear the vehicle is driven, and whether all the
wheels are coupled.
SUMMARY OF THE INVENTION
It is the object of the invention to devise a method of controlling the
brake pressure in an anti-lock vehicle brake system which will provide a
short distance to stop, at good steerability of the vehicle, even if the
mass moments of inertia vary greatly. The method is to provide good
braking performance both at low and good coefficients of friction as well
as with quick changes between loading and unloading of the wheel.
Under-braking is to be prevented.
The solution according to the invention to meet that object provides to
calculate the pressure reduction period of a control cycle currently under
way as the product of a first factor which depends on the pressure
reduction period in the preceding control cycle and at least a second
factor which depends on the maximum re-acceleration in the preceding
control cycle of the wheel.
The pressure reduction period in the very first control cycle (for which
there is no preceding control cycle) is determined and measured in
conventional manner by slip and/or deceleration threshold values being
exceeded with respect to the braked wheel. The value measured for the
pressure reduction period is stored in the computer and serves as the
basis for dimensioning the pressure reduction period in the next control
cycle.
In accordance with a preferred and especially simple embodiment of the
invention the first factor is the pressure reduction period in the
preceding control cycle itself. As an alternative to this solution, the
first factor may be determined quite generally in response to the pressure
reduction period in the preceding control cycle, e.g. by having this
pressure reduction period of the preceding control cycle, multiplied by a
certain multiplicator (weight factor), included in the final result for
the pressure reduction period of the control cycle currently under way.
In a preferred simple modification of the invention the second factor which
is dependent on the maximum re-acceleration of the wheel in the preceding
control cycle results as the quotient of a given value and the maximum
re-acceleration of the wheel measured in the preceding control cycle.
Thus the pressure reduction period according to the invention results in
accordance with the following equation:
##EQU1##
wherein
t.sup.n.sub.ab is the pressure reduction period in the control cycle (n)
currently under way
t.sup.n-1.sub.ab is the pressure reduction period in the preceding control
cycle (n-1)
peak-soll is a predetermined value (if desired, as a function of certain
parameters, see below) and
peak.sub.n-1 is the maximum re-acceleration of the braked wheel in the
preceding control cycle (n-1).
Thus the pressure reduction period for a control cycle currently under way
always is calculated anew, according to the invention. The starting base
for that calculation are data which were measured in the preceding control
cycle, namely on the one hand the pressure reduction period in the
preceding control cycle and, on the other hand, the maximum
re-acceleration of the wheel in the preceding control cycle. If the
maximum re-acceleration in the preceding control cycle was relatively
small, then the pressure reduction period in the control cycle currently
under way becomes relatively great compared to the pressure reduction
period in the preceding control cycle. Inversely, if the re-acceleration
of the wheel in the preceding control cycle was rather great, then the
pressure reduction period in the control cycle currently under way becomes
rather small compared to the pressure reduction period in the preceding
control cycle.
The above describes the basics and the principal control parameters of the
braking algorithm. Under certain circumstances the invention provides for
refinement and modifications of the basic algorithm according to preferred
embodiments. According to a preferred variant of the invention, for
instance, the pressure reduction period of a control cycle currently under
way is extended as compared to the time period resulting from the above
mentioned two factors if the deceleration and/or slip of the braked wheel
become greater than given threshold values. The fact that the
predetermined threshold values are exceeded is an indication of the risk
of the wheel running into a particularly unstable state so that it is
advantageous to extend the pressure reduction period in the instantaneous
cycle. The degree of deceleration and/or slip of the wheel may be measured
and the extension of the pressure reduction period effected in proportion
to the measurement values. For example, the pressure reduction period may
be extended by one millisecond per 2 km/h slip or per 2 g wheel
circumference deceleration, each above a predetermined threshold value. In
this manner rapid pressure lowering is accomplished in case of a so-called
negative .mu. jump (leap in coefficient of friction).
As explained above, in accordance with the basic algorithm of the
invention, the pressure reduction period in the control cycle currently
under way results as a product of two factors, the second factor being a
quotient which has in its denominator the maximum re-acceleration of the
wheel measured in the preceding control cycle. In the numerator of the
quotient there is a value which is fixedly given according to a most
simple embodiment of the invention. According to a preferred further
development of the invention the given value is varied as a function of
the reduction in speed of the wheel during the preceding control cycle.
If, in the preceding control cycle, there was only little decrease in
speed after pressure reduction had begun that is an indication that the
curve of coefficient of friction/slip continues to rise (e.g. deep snow)
so that the decreases in pressure (i.e. pressure reduction periods) should
be relatively small. If the decrease in velocity measured in the preceding
control cycle is small, therefore, the given value used in the numerator
of the quotient is kept relatively small in order to achieve a brief
pressure reduction period.
When the pressure reduction period calculated according to the invention
for the control cycle currently under way has expired, the pressure
reduction is interrupted and the brake pressure at the brake of the wheel
concerned is kept constant for the time being. According to a preferred
embodiment of the invention the pressure reduction is continued when the
wheel circumference speed has diminished by a certain amount of e.g. 2 to
10 km/h since termination of the preceding pressure reduction. Of course,
that applies only if the wheel is not yet in a condition where the threat
of locking is imminent. In other words: when the wheel gets into a state
of impending locking (found out conventionally by measuring the
deceleration and/or slip and comparing it with threshold values) keeping
the pressure constant is interrupted at once and the reduction of the
pressure is continued. The pressure is not maintained constant when there
is a tendency of locking.
According to another modification of the method according to the invention,
on the other hand, the pressure reduction is terminated at once if the
braked wheel experiences re-acceleration greater than a predetermined
threshold value during the instantaneous control cycle prior to the end of
the pressure reduction period.
A further refinement of the method according to the invention takes into
account that the maximum re-acceleration of the braked wheel measured in
the preceding control cycle does not mean much for the renewed calculation
of the pressure reduction period if the so-called control deviation
(pressure variation p during the control cycle) was very small in the
preceding control cycle. In that event, therefore, the pressure reduction
period in the instantaneous control cycle is equalled to the pressure
reduction period of the preceding control cycle, and the above
modifications provided under special circumstances to the pressure
reduction period are applicable as well.
The algorithms described above for control of the brake pressure in an ABS
brake system are implemented by software in line with present day
technology, i.e. by the programming of a processor. Those skilled in the
art nowadays are familiar also with the means by which to measure and
process the various control parameters.
BRIEF DESCRIPTION OF THE DRAWINGS
An embodiment of the invention will be described further with reference to
the drawing, in which:
FIGS. 1a, 1b, and 1c diagrammatically show the course of the
circumferential speed of a braked wheel, with the pressure reduction
period once being set too long, once too short, and one correctly, plotted
above a common time scale, including the corresponding courses of the
wheel acceleration a.sub.rad and the corresponding courses of the brake
pressure p.sub.rad at the braked wheel, and
FIGS. 2a and 2b are flowcharts of a braking algorithm according to the
invention.
DESCRIPTION OF THE PREFERRED EMBODIMENT
In FIG. 1 three functions are plotted above a common time scale (abscissa).
FIG. 1a shows the course of the circumferential speed v.sub.Rad of a
braked wheel and the per se known reference speed v.sub.Ref. FIG. 1b shows
the course of the acceleration of the wheel, including a desired
theoretical value, above the same time scale, and FIG. 1c shows the
corresponding course of the brake pressure, again above the same time
scale.
As designated in FIG. 1a, the pressure reduction period is too long at the
first minimum of the wheel circumference speed; the pressure is reduced
too much. Accordingly, in FIG. 1b the re-acceleration of the wheel reaches
a relatively high maximum value I which is clearly above the theoretical
value S.
The next minimum of the wheel circumference speed v.sub.Rad in FIG. 1a
corresponds to a pressure reduction period which is too short, i.e.
insufficient pressure reduction. Accordingly, the re-acceleration reaches
a relatively flat maximum II which lies below the theoretical value S.
It is only at the minimum shown at the far right in FIG. 1a of the wheel
circumference speed v.sub.Rad that the pressure reduction period and,
therefore, also the lowering of the pressure are correct so that the
maximum III of the re-acceleration curve coincides with the theoretical
value S (FIG. 1b).
It becomes clear from FIG. 1 that the re-acceleration measured in a control
cycle permits an observation to be made as to whether the pressure
reduction period was too short or too long.
FIGS. 2a and 2b present a flowchart according to which the braking
algorithm described above can be implemented.
Top